FORN VÄNNEN JOURNAL OF SWEDISH ANTIQUARIAN RESEARCH

2020/1 Utgiven av Kungl. Vitterhets Historie och Antikvitets Akademien i samarbete med Historiska museet.

Fornvännen finns på webben i sin helhet från första årgången och publiceras löpande där med ett halvårs fördröjning: fornvannen.se

Ansvarig utgivare och huvudredaktör Mats Roslund Vitterhetsakademien Box 5622, 114 86 Stockholm [email protected]

Redaktionssekreterare och mottagare av manuskript Peter Carelli Vitterhetsakademien Box 5622, 114 86 Stockholm [email protected]

Redaktörer Herman Bengtsson, [email protected] Christina Fredengren, [email protected] Åsa M Larsson, [email protected]

Teknisk redaktör Kerstin Öström Grävlingsvägen 50 167 56 Bromma [email protected]

Prenumeration Vitterhetsakademien Box 5622, 114 86 Stockholm e-post [email protected] Bankgiro 535-3552

Årsprenumeration i Sverige (4 häften) 200 kronor, lösnummer 60 kronor

Journal of Swedish Antiquarian Research published by The Royal Academy of Letters, History and Antiquities Subscription price outside Sweden (four issues) SEK 250:– Box 5622, SE-114 86 Stockholm, Sweden fornvännen började utges av Kungl. Vitterhets Historie och Antikvitets Akademien år 1906 och ersatte då Akademiens Månadsblad samt Svenska Fornminnesföreningens Tidskrift, som båda tillkommit under 1870-talets första år. Förutom i Sverige finns Fornvännen på drygt 350 bibliotek och vetenskapliga institutioner i mer än 40 länder. Tidskriften är referentgranskad. fornvännen (»The Antiquarian») has been published by the Royal Academy of Letters, History and Antiquities since 1906, when it replaced two older journals which had started in the early years of the 1870s. Outside Sweden Fornvännen is held by more than 350 libraries and scientific institutions in over 40 countries. The journal is peer-reviewed. issn 0015-7813 Printed in Sweden by AMO-tryck AB, Solna, 2020 Författarinstruktioner

fornvännen välkomnar manuskript i nordisk arkeologi och äldre tiders konstvetenskap med angränsande ämnen. Bidrag skrivs på ett av de skandinaviska språken (svenska, norska, danska) eller på engelska. Fornvännen utkommer med fyra tryckta nummer per år och innehållet tillgängliggörs Open Access med ett halvårs fördröjning.

fornvännen tar emot fyra typer av bidrag: 1) Artiklar: Fördjupade presentationer av nya forskningsresultat som offentliggörs för första gången. Max 45 000 nedslag inkl. blanksteg (exkl. abstract och referenser). Kompletteras med abstract, referenser, summary och 4–5 keywords på engelska.

2) Korta meddelanden: Korta presentationer av fynd/resultat som bedöms ha forskningsmässigt nyhetsvärde. Max 13 500 nedslag inkl. blanksteg (exkl. ev. referenser). Får ha ett begränsat antal referenser och illustra- tioner. Ska ej ha abstract eller summary.

3) Debattinlägg: Kortare diskuterande och argumenterande texter. Max 20 000 nedslag inkl. blanksteg. Får ha ett begränsat antal referenser och illustrationer. Ska ej ha abstract eller summary.

4) Recensioner: Granskning av både skandinaviska och internationella publikationer som bedöms vara av intresse för tidskriftens ämnesområden. Max 9 000 nedslag inkl. blanksteg. Inga illustrationer eller refereser.

Alla bidrag granskas av Fornvännens redaktion samt en extern sakkunnig genom s.k. dubbel blind peer review.

Utförliga författarinstruktioner för innehåll, format, referenssystem osv. finns att ladda ned på: http://vitterhetsakad.se/publikationer/fornvannen. Bidrag tas emot löpande och ska skickas i färdigt skick, redigerade enligt de utförliga författarinstruktionerna, och med illustrationer till: [email protected].

Guide for Authors

fornvännen: Journal of Swedish Antiquarian Research accepts manuscripts in English on Nordic archaeology and early art history, as well as topics closely related to these disciplines. The journal is published in four printed issues annually, and the contents are made Open Access six months after publication.

Four types of contributions can be submitted: 1) Articles: Max. 45,000 characters incl. spaces (excl. abstract and references). Abstract, summary and 4–5 keywords in English.

2) Short Communications: Brief presentation of artefacts/results judged to be newsworthy for researchers. Max. 13,500 characters incl. spaces (excl. references). A limited number of illustrations and references allowed, but no abstract.

3) Debates: Max. 20,000 characters incl. spaces (excl. references). A limited number of illustrations and references allowed, but no abstract.

4) Reviews: Scandinavian and international publications judged to be of interest to the readership of the journal. Max. 9,000 characters incl. spaces. No illustrations or references.

All submissions undergo editorial auditing and revision. Article manuscripts are subjected to double-blind peer review by external researchers.

Detailed instructions for authors covering content, format, reference system etc. can be downloaded at: http://vitterhetsakad.se/publikationer/fornvannen. Manuscripts, completed according to the full instructions and including illustrations, may be submitted at any time to: [email protected]. Innehåll

1 Isberg, A. Ringspännen i Lund. Skiftande recensioner traditioner mellan Östersjön och Konti- 61 Ling, M. & Welinder, S. Arkeologiska fan- nenten. —Summary. tasier. Anmälan av J. Wienberg. 15 Ødeby, K. Hvor er runene? —Summary. 63 Dørum, K. & Holberg, E. Frå høvdingdøme 29 Lingström, M. & Klackenberg, H. Sigill- til statsmakt i Noreg ca. 200–1350. An- stampen från Biskops-Arnö. —Summary. mälan av T. Lindkvist. 65 Skoglund, P. et al. (red.). North Meets South: debatt Theoretical Aspects on the Northern and 37 Hegardt, J. Kanske: Det humanistiska Southern Rock Art Traditions in Scan- tvivlet och det arkeologiska narrativet. dinavia. Anmälan av V. Mantere. 40 Björklund, A. et al. Referensverket Det 67 Zachrisson, I. Gränsland: Mitt arkeologiska medeltida Sverige och arkeologin. liv. Anmälan av I. Berg. 43 Indrelid, S. Medieval communal reindeer 69 Buchwald, V.F. Jernets Danmarkshistorie. drives on Hardangervidda, South Nor- Anmälan av A. Ödman. way: reply to Otto Blehr.

korta meddelanden 51 Oehrl, S. Gotland’s largest picture stone rediscovered. 55 Gustafsson, N.B. Ett nyfunnet märke från Roma kloster. 58 Larsson, A. et al. Korsvikskäken: Ett bohuslänskt skelettfynd från mellersta järnålder.

issn 0015-7813 Debatt 43 Medieval communal reindeer drives on Hardangervidda, South Norway: reply to Otto Blehr

In his two debating articles in Fornvännen (2012 excavated at Sumtangen and Store Krækkja in and 2019), Otto Blehr discusses, and strongly cri- the first half of the 1970’s. Before this, starting as ticizes, my work on medieval reindeer hunting on early as 1840, several other researchers excavated the Hardangervidda. As his allegations are essen- house foundations and middens (Indrelid 2014). tially the same in both contributions, I choose to Professor Hufthammer and Dr. Liselotte Tak- respond to both simultaneously, and to discuss ken-Beijersbergen analysed the faunal material his main arguments here. from the excavations 2004–2008 osteologically In cooperation with colleagues from within and statistically. DNA-analyses were done by the and outside the field of archaeology, I have re- geneticist, professor Knut Røed, and the archae- searched the large scale, but seemingly brief mass ologist Dr. Jostein Bergstøl has registered cairns hunting of reindeer in the northern part of the and postholes that may have been connected to mountain plateau Hardangervidda in Norway in the hunt. medieval times (Indrelid 2010; 2013; 2014; 2015; Blehr’s criticism concentrates mainly on the Indrelid & Hufthammer 2011, Indrelid et al. following issues: 2007; 2015). The background for this is partly a) The traditional legend and my use of it. provided by a local legend concerning a hunting b) The remains of cairns and other leading technique making use of leading fences and float- contraptions in the terrain. ing lines on water guiding animals into lakes to c) When did the communal drives at Sum- be killed, and partly by material from archaeo- tangen take place, and how long did they logical excavations of nearby stone hut ruins and last? bone middens. d) How extensive was the hunt? Sumtangen, a strip of land on the southern e) Who organised the communal reindeer shore of Lake Finnsbergvatn (1190 m a.s.l.), is drives? one of the most important locations for the leg- end as well as for cultural remains connected to a) The legends the legends. Two stone hut ruins, named Austbu The legends describing reindeer hunting connec- and Vestbu, lie near the narrow sound where the ted to four lakes on the Hardangervidda mountain hunt is said to have taken place. They are sur- plateau, Finnsbergvatn, Store Krækkja, Ørteren rounded by large bone middens, consisting al- and Nordmannslågen, are fragmentary and partly most entirely of reindeer bones and antler frag- self-contradictory. However, they contain some ments. common elements such as catching reindeer flocks As a part of an interdisciplinary project led by in water, rows of cairns that lead the animals the osteologist, professor Anne Karin Huftham- towards the lakes, and the use of ropes attached mer and me, small parts of the waste middens at to the cairns or stretched over water. All four Sumtangen were excavated in the years 2004– hunting stations feature large stone hut ruins 2005, and a further two comparable archaeologi- surrounded by middens consisting of reindeer cal sites ca. 15 km further south on the shore of bones. As far as I know, there are no other mid- Lake Store Krækkja in 2006–2008. In 2006, dens of comparable size in the northern or cen- assisted by marine archaeologists, we collected tral part of the Hardangervidda. bone material from a fourth, now inundated, The manuscript collection of the University hunting station in Lake Ørteren, 4 km northeast library in Bergen contains several notes by the of Store Krækkja. regional official, stiftamtmann W.F.K. Christie, In addition to having discovered leading fen- concerning cultural heritage sites connected to ces and trapping pits in the landscape, Otto Blehr the reindeer hunt on Hardangervidda (Christie

Fornvännen 115 (2020) 44 Debatt ms. 222 nr. XIII, XIV and XV). These state that he gen’s Museum. His account of the legend about learned of house foundations and bone middens Sumtangen is consistent with Nicolaysen’s ver- on Hardangervidda in 1838. The collection further sion (Bendixen 1892, p. 27). contains report of a journey undertaken in 1840 by There are two hunting stations with remains W. Koren, who, on Christie’s request, visited and of stone huts and middens on the southern shore examined these and other sites on Hardangervidda. of Lake Store Krækkja (Nordre Boplass and Sønd- Koren’s report was reprinted and transcribed in re Boplass), whilst the remains of two large stone extenso by Øivind Bakke (Bakke 1985). hut ruins and a bone midden are situated at Lake Koren’s report describes the hunting drives Ørteren. The latter was dammed and inundated on Hardangervidda for the first time, although in the 1960’s. These three stations are connected not from Lake Finnsbergvatn, but from Nord- to a legend concerning a complex of hunting mannslågen, a lake situated ca. 30 km further structures near the narrow isthmus Svoi, which south. He states that reindeer were caught using divides Store Krækkja from Ørteren. Bendixen ropes stretched over the water, whilst they were was told the legend while visiting Krækkja in trying to swim across the lake from the opposite 1890. Ropes were supposedly set up at Svoi, reach- shore. In a letter to Christie, dated 20 November ing into the lake where they formed a curve. “Når 1840, he explains that this legend also concerns rensdyrene så ikke kunne komme frem over eidet the bone middens in what he refers to as Findse- og satte på svøm, stødte de overalt paa linerne og dalen (at Finnsbergvatn). In a note (Ms. 222 nr. blev overfaldne med spyd og pile” (When the XIII), probably written in the autumn of the reindeer could not cross the isthmus and took to same year, Christie mentions the same fragment the water, they would meet the ropes everywhere of the legend and adds: “Der siges at dyrene ere and were ambushed by spears and arrows, my fangede paa vandet med liner af beboerne” (It is translation) (Bendixen 1892, p. 35). said that the animals are caught with ropes in the Around twenty years later, Yngvar Nielsen, a water by the inhabitants, my translation). This professor in geography, refers to the same legend, concerns the house foundations and bone mid- adding that old cairns can be found at Svoi, to dens at Sumtangen. Thus, it is not correct when which, according to the legend, ropes were fas- Blehr in his summary of the 2019 article claims tened that continued into Lake Krækkja by that Christie did not mention any legend con- which the reindeer would get trapped and were cerning Sumtangen. killed (Nielsen 1909, p. 593). The antiquarian Nicolay Nicolaysen, who Fragments of legends that survived the 600 visited Sumtangen in 1860 (Nicolaysen 1861), to 700 years since the hunt took place should was told that ropes were stretched between two hardly be taken literally. After being retold by 15 promontories in Lake Finnsbergvatn (the south- to 20 generations, not much can be left of the eye- ern of these being Sumtangen), trapping swim- witness accounts describing the original events. ming animals in this way. Therefore, I am not as concerned by the details in In 1884 Th. S. Haukenes published a more the individual fragments as Blehr seems to be. I comprehensive legend linking the drives to trap- see a connection between the hunts at all four ping pits near Gravskar, 2.5 km west of Sumtan- stations that, according to their 14C-dates, were gen. Rows of cairns, connected by ropes with ratt- in simultaneous use. The individual elements in ling bells on them, would lead the reindeer to- the fragmentary legends can at best provide some wards the trapping pits and the lake. Some were suggestions of how the hunt may have happened. caught in the pits; others trying to escape through the lake were killed by hunters in boats. This b) The remains of cairns and other leading obviously is a combination of fragments of two contraptions in the terrain legends, one from Gravskar and another from In the 1970’s, Blehr described six cairns on the Sumtangen (Bakke 1985, p. 100). north side of Finnsbergvatn, claiming these had In 1890 and 1891 Bendixen, a school director, nothing to do with Sumtangen, because they lie visited Hardangervidda on a mission from Ber- further to the west where the lake is widest. He

Fornvännen 115 (2020) Debatt 45 may be right there, but he is mistaken when he the onset of the rutting season have consisted of thinks that these cairns are included in the rows only a few animals” (Blehr 2012, p. 119). Accord- of cairns and post holes I mentioned in 2010 (p. ing to Dr. Terje Skogland, one of the most distin- 32), and by which he expresses his dissatisfaction guished biologists in international reindeer sci- in his 2019 article. Dr. Bergstøl, a participant in ence, about 500 bulls and a herd of several hun- our project, proved the existence of several rows dreds of cows were counted within Hardanger- of cairns and postholes north of Lake Finnsberg- vidda North (including the Finnsbergvatn area) vatn, opposite Sumtangen, in 2008–2010 (Berg- in 1985. He does not give the impression that støl 2000, fig. 3, p. 102): “At the top of the east- finding bull herds and fostering flocks in this area ern row, there are several post holes that show is unusual. This is in accordance with the con- that it [the row] must have been reinforced at this tents of the bone middens at Sumtangen that spot. The western row follows the ridge that ends indicate that bulls, cows, and calves were target- just above Sumtangen. The eastern row prevents ed in the hunts. It does not necessarily mean that the animals from moving to the eastern side of these were part of the same group. I believe that, Lake Finnsbergvatn. It is difficult to determine during late summer and autumn, the hunters whether the animals entered the lake voluntarily stayed at Sumtangen so long as there was any or whether they were driven into it.” (Bergstøl chance of capturing flocks of profitable size. 2000, pp. 106–107, my translation). In an e-mail Sometimes these would consist of bulls; some- to me, dated 15.11.2019, Dr. Bergstøl describes times these would be fostering flocks. these more accurately as being small cairns and At Sumtangen, there is also a midden from several dozens of postholes. These contraptions the Older Iron Age, ca. AD 200–400. According can hardly have had another purpose than to lead to professor Knut Røed, the geneticist who per- animals towards the shore, barely 200 m from formed DNA-analyses on the bones from the Sumtangen on the opposite side of the lake. Older Iron Age, medieval period, and the pres- Blehr seems to accept that some form or other ent-day Hardangervidda reindeer population, of “communal reindeer drive” occurred at Sum- the genetic variation was stable until the Middle tangen. Somehow, he has conceived the idea that Ages, but “there has been a substantial change the hunt targeted bulls exclusively. In his 2012 from the Middle Ages until the present” (Indre- article, he accounts for an analysis of antler re- lid et al. 2007, pp 151–152). “The Hardangervidda mains he recovered from the Sumtangen midden reindeer population has gone through rather dra- in 1972. He cites the zoologist Odd Kjos-Hans- matic genetic deteriorations, especially during sen who states that “Both sexes and all age groups the last two centuries, when reindeer husbandry were represented in the material, but with a was practised in this mountain region” (Røed et slight overweight for bull antlers” (Kjos-Hans- al. 2011). Furthermore, it is known that reindeer sen 1973, p. 77–78), but is like Kjos-Hanssen sur- migration routes are highly influenced by human prised by this result: “Since such a male domi- infrastructure and disturbance. On Hardanger- nance is hardly likely in an area used by fostering vidda, reindeer migration patterns are known to flocks, where there would have been an excess of have changed a.o. after the establishment of the females and calves, this indicates that the area railroad between Oslo and Bergen (Skogland & was mainly visited by bulls.” He concludes: “Thus, Jordhøy 1988). Therefore, drawing definite con- in most cases it would have been bulls migrating clusions on reindeer migration routes in medie- south in the fall that were driven into the water.” val times based on observations of presentsday (Blehr 2012, p. 119). reindeer can only be speculation. Referring to his own observations of reindeer over a period of four years in the Sumtangen c) When did the communal drives at Sumtangen area, he informs that the largest flock he ever take place, and how long did they last? encountered were roughly twenty bulls. These In our 2011 paper we presented the 15 radiocar- data are of importance, he says, “since they indi- bon dates of bone for the medieval middens at cate that most of the flocks migrating south at Sumtangen we so far had received (specified in

Fornvännen 115 (2020) 46 Debatt Blehr 2012, fig. 1). The mean date interval at a 68 reject the idea that large numbers of reindeer were per cent confidence level showed the period AD caught at Sumtangen. He claims that “herd” is a 1240–1290 (Indrelid & Hufthammer 2011, p. concept used to describe “a larger unit within a 48). Since then, three new bone samples from the population, sometimes for post-calving aggrega- stratigraphically oldest bone layers have been tions of 1000–5000 animals” and refers to Parker dated (Takken-Beijersbergen 2017, p. 343 and (1972). Parker’s article concerns barren-ground tab. 2). caribou in north-central Canada. I have no know- Based on the median probability values, 12 of ledge of Canadian caribou group size, but both the 18 radiocarbon dates lie between AD 1215 Blehr and I know that reindeer flocks on Hardan- and AD 1300, with an average of AD 1257. Five gervidda in our times can vary between a few lie between AD 1096 and AD 1193, with an aver- dozen or fewer individuals to several hundreds, age of AD 1155, and one AD 1348. The dates indi- occasionally even more than 1000 animals. cate a main hunting period restricted to the 13th Because the estimates professor Huft-hammer century, but with a preceding phase in the early and I provide for the average yearly catch are con- or middle part of the 12th century. It is unclear siderably lower than 1000–5000, Blehr con- whether these older dates are related to the mass cludes audaciously: “Thus, it goes without saying hunt. Six dates from three smaller middens at that no «mass killing of entire herds» can ever Lake Krækkja, not connected with stone hut have taken place at Sumtangen.” (Blehr 2019, p. remains, have been dated to the period AD 1096– 243). 1185. These sites contain the remains of but a few We have no objections to replace “herds” by bulls and are most probably unrelated to the mass “flocks” or other terms that Blehr would find hunts. more appropriate. Our point is that large groups Based on three radiocarbon dates from his of animals were driven into the water, dragged own excavation in 1973, Blehr concluded that onshore and butchered by the huntsmen. The “the hunt had been carried out regularly through remains of these, by our tentative estimations, the Early Middle Ages only to end abruptly with 5500–7800 animals, are left behind in the bone the Black Death 1349–50” (Blehr 2012, p. 116). midden at Sumtangen. In my opinion, this justi- On the following page, he presents the 15 radio- fies using the term “mass killing”. Concerning the carbon dates from our excavations, and admits extent of the hunt, Blehr refers to Dr. Takken- that “Judging from them, it is likely that the com- Beijersbergen who according to him reaches munal drives at Sumtangen ended already in the completely different conclusions. Different sta- early 14th century, that is, before the Black Death tistical methods may be used to estimate the hit the area.” Despite this, Blehr chooses to trust number of individuals. We used MNI (minimum his own three dates from 1973 more (740±80, number of individuals) (White 1953) and MLNI 600±100 and 700±100 BP) and say that these, (most likely number of individuals) as described “when calibrated, are consistent with an end date by Adams & Koningsberg (2004). of about 1350, though with so few dates and such Our material was based on the bone material low precision it could have been a century later from one square meter, 0.5 m3 of sediment and too” (Blehr 2012, p. 121). Blehr’s opinions on the bones excavated from the midden on the south excellence of his own ideas are at times astound- side of the Austbu ruin, in total 22.749 bones and ing. bone fragments (Indrelid & Hufthammer 2011, tab. 2). The whole midden outside Austbu and d) How extensive was the hunt? Vestbuis estimated to 50 m3 (Indrelid et al. 2007, In his 2019 article Blehr criticizes our use of the p. 137). terms “herds”, describing the reindeer groups that The MNI method, based on 55 right carpale were caught at Lake Finnsbergvatn, and “mass 4+5, gave the result of 5500 individuals. The killing”, describing the hunting method (Indre- MLNI estimate derived from the calcaneum gave lid & Hufthammer 2011). He allows his ponder- 5800, from the astragalus 7800, and the centro- ing on these words to become an argument to tarsale 7700 individuals. Dr. Takken-Beijersber-

Fornvännen 115 (2020) Debatt 47 gen reaches a lower estimate. She concludes that that it was not by townspeople. In his 2012 article the middens contain the remains of a minimum however, he presents a “tentative hypothesis”, of 2851 animals based on an MNI of 77. This is saying that he can see a connection between the considerably less than the MNI of 5500 and indi- communal reindeer drives at Sumtangen and the cates that the bones were not distributed evenly iron production in Sysendalen, especially at the in the middens. However, one has to assume that farm Fet: “The production of iron, and the char- more animals than calculated were represented coal this process demanded, would have emp- at the site, because of the very nature of the MNI, loyed many people, and thus made the meat from and the likely size reduction of the middens due reindeer killed at Sumtangen a necessary supple- to taphonomic processes, including later human ment to the food produced at Fet.” (Blehr 2012, disturbances and the earlier excavations. Both p. 120). An observation made by Nicolaysen (1861, the archaeological evidence and the demographic p. 18), that some of the charcoal kilns were not composition of the taphocoenoses strongly indi- emptied, made him propose that something had cate a hunting strategy that was aimed at killing happened to their owners, and suggests that it as many animals as possible (Takken-Beijersber- had to do with the Black Death. Blehr finds this gen 2017, p. 353). to be “a reasonable assumption” and goes on BlehrmakesapointofthefactthatDr.Takken - speculating: “If the plague killed the men burn- Beijersbergen’s results show lower numbers than ing charcoal, it would also have killed enough of ours and takes this as an evidence that no mass those engaged in iron production as well as those hunt can ever have taken place (2019 p. 243). It hunting at Sumtangen.” Referring to his own should be self-evident that no statistical method three radiocarbon dates from 1973, he concludes: ever can give exact numbers of the real catch. Our “Thus, it seems certain that the communal rein- objective was to find out whether the total catch deer drives at Sumtangen did come to an abrupt was large – several thousand, or less extensive. end when the plague hit the area in 1349” (Blehr The results demonstrate beyond a reasonable 2012, p. 121). doubt that we are not talking of a few hundred In my opinion, there are three facts that indi- individuals but several thousands. cate that the hunt was organised by others than In his comments (2019, pp. 242–243), he ob- local huntsmen: 1) Laws, and finds from the mid- viously regards our average numbers as absolute. dens; 2) Runic inscriptions and rare artifacts; 3) As an example, we calculated the annual catch The way the animals were butchered. over a 50 years period to be 110 individuals using I have, in several publications, discussed legal 5500 animals as a total number (Indrelid & texts concerning the ownership of swimming Hufthammer 2011). This is of course a hypothet- animals that are killed (Indrelid 2010; 2014; ical example. Some years the number would be 2015). Both the Gulating Law Codex, one of the considerably higher, some years lower. When the four regional laws in Norway in the Early Middle catch for several years was lower than what was Ages, and the first national law (Landsloven 1274) profitable, the communal hunt would be aban- contain rules about this. The Gulating Law re- doned, which according to our dates happened gion covered a large part of South Norway, in- several decades before the Black Death 1349– cluding the districts of Hardanger to the west, 1350. and Hallingdal to the east of Hardangervidda. Part V, article 24, describes a particular situation e) Who organised the communal reindeer drives? when a swimming animal is killed. If the one who In his 2019 contribution, Blehr states that “There kills the animal is someone else than the person is no evidence that townspeople and not the who “owns” it, the upper foreleg (bóg) belongs to hunters themselves organized the communal dri- the one who killed it. In the law text, this is ves in medieval times.” (Blehr 2019:245). In this referred to as skot-bóg. case, Blehr and I have interpreted the data differ- As mentioned below, precisely upper forelegs ently. I cannot see that he expresses a clear theory are heavily underrepresented at Sumtangen. I of who organised the communal hunt, except consider it likely that the upper foreleg was the

Fornvännen 115 (2020) 48 Debatt hunters’ payment, whilst the rest of the carcass inscriptions are known from Hallingdal, 11 of fell to the “owners”. The “owner” of the animal those are from Torpo church and four from the in this case should be interpreted as the one who church in Ål. There are five inscriptions from organised the hunt or was considered the owner Hardanger: one from Odda, one from Kvam and of the game in the area. The king considered him- three from Ullensvang. Based on this, it is douptful self the owner of all the land that fell outside the that literacy was widely spread amongst the rural area of the established farms, but the local people population in Hallingdal and Hardanger in the had some hunting and fishing rights in such Middle Ages. areas. Reindeer hunting to such an extent would There is, however, more than just the runic hardly have been accepted by the king. There- inscriptions alone that make me doubt that it fore, I consider it highly unlikely that local hun- were local reindeer hunters that organised the ters organised the communal reindeer hunts on communal hunt on Hardangervidda. Some oth- Hardangervidda in medieval times. er finds also seem strange amongst reindeer hun- I have posed this question to Dr. Jørn Øyre- ters in the field. A chess piece was found in one of hagen Sunde, professor in legal history and spe- the middens at Sumtangen, two dices for game at cialised in medieval sources. He agrees with my Krækkja North, and at Krækkja South a frag- arguments. On the Landsloven that succeeded ment of a long bone featuring small, carved circ- the Gulating Law, he says a.o.: “Combining all les identical to those that were used to decorate the rules in the Landsloven from 1274, there can medieval combs. This is an ordinary bone frag- be no doubt that the king prevailed over all acti- ment on which somebody played around with a vity in commons (“allmenninger”) that exceed- pair of compasses. In my view, these artifacts do ed established use.” The extensive, but relatively not fit in with the equipment of a reindeer hun- brief communal reindeer hunt he interprets as “an ter, but they may well have belonged to merchants example of use of such a character and where the or craftsmen present during the hunt to attend to timing of its origin that the king must approve it. their business- and crafting interests. The person to grant permission for such a hunt Blehr omits mentioning one of the most im- would have been the “sysselmann”, the king's portant arguments I have put forward against a highest regional official” (my translation) (Øyre- locally organised hunt: the way the animals were hagen Sunde, e-mail to me 24.11.2019). butchered. All skeletal elements are present in In his 2019 contribution, Blehr mentions four the 13th century middens at Sumtangen, except reindeer bones with runic inscriptions found in the main part of the antlers, ribs and upper fore- the middens at Sumtangen. I consider it more leg bones. Marrow-containing bones are heavily likely that these runes were carved by other people fragmented. The meat was cut from the bones than the reindeer hunters in Hallingdal and Har- during the butchering process, and almost all danger. Blehr refers to the Samnordisk runtext- bones were left as waste at the site. This is com- databas and states it as a fact that “medieval runes pletely different from the butchering pattern were in common use at least to ca. 1400 for sim- known from the local communities around Har- ple messages inscribed on wood and bone”. That dangervidda in historical time, where the head runes were used regularly in certain circles in the and distal parts of the feet were chopped off, towns, is commonly accepted. Carvings from a.o. whilst the remainder of the carcass would be churches in the countryside prove that even some transported in one piece or divided into shoul- villagers must have been familiar with the runic ders, legs, loin and ribs. Why did those who under- alphabet. However, we know little about how took the mass-hunts at Sumtangen and Store common this knowledge was. Krækkja treat the carcasses differently? Professor Henrik Williams at the Department Bearing in mind the large amounts of meat to of Scandinavian Languages of Uppsala University be carried away after a successful mass-hunt, it kindly sent me an overview over all medieval was necessary, even with horse transportation, to runic inscriptions from Norwegian rural con- keep the weight at a minimum. If the destination texts in the Samnordisk runtextdatabas. Fifteen for the products was far away, it would have been

Fornvännen 115 (2020) Debatt 49 worth the bother of cutting off the meat and leav- I find it lamentable and sad that Blehr lets his sci- ing most of the bones behind, that only present- entific legacy be oveshadowed by this type of ed unnecessary weight and ballast. All marrow- one-sided negative narrative of the works of a containing bones were shattered, either to con- former colleague and friend. I agree wholeheart- sume the nutritious marrow on site or to trans- edly with him that “As researchers we have a port it with the meat. If the destination was responsibility to present a picture as solidly under- close-by, like e.g. to Fet in Sysendalen, like Blehr pinned as possible, based on our limited data” suggested, it seems absurd that the hunters would (Blehr 2019, p. 245) and leave it to the reader to debone the animals first, harvest the marrow and judge who provides “unsubstantiated statements”. then undertake the 25 km long trip home to Fet. This discussion is closed as far as I am con- The antlers were taken away from site as well. cerned. The middens contain almost exclusively smaller tines and skulls where the antlers have been re- Acknowledgement moved. Reindeer antlers were an important com- I would like to express my sincere gratitude to Dr. modity in medieval times. They were used for Liselotte Takken-Beijersbergen for significant a.o. the production of hair combs and other per- assistance during my work with this debate con- sonal accessories. Combs made of reindeer antler tribution. are found in medieval archaeological contexts spanning from Northern Norway to Schleswig and from Lund to Orkney and the Faroe islands References (Røed & Hansen 2015, pp. 69–70). The remains Adams, B. J. & Koningsberg, L. W., 2004. Estimation of a comb maker's workshop were found in the of the most likely Number of Individuals from commingled Human Skeleton Remains. American medieval layers at Bryggen in Bergen, featuring Journal of Physical Anthropology, 125:138–151. debris of reindeer antler off cuts (Herteig 1969, Bakke, Ø., 1985. W. F. Korens innberetning fra Har- p. 186). dangervidda i 1840: Med innledning og kom- The way the animals were butchered is the mentarer av Øivind Bakke. Hardanger, 1985:54– most convincing argument for that not local 116. huntsmen were responsible for hunt, and that Bendixen, B. E., 1892. Fornlevninger i Hardanger. Foreningen til Norske Fortidsmindesmerkers Bevaring: the resulting products were not intended for the Aarsberetning for 1891. Kristiania. local villages, but more likely for the towns, in Bergstøl, J., 2000. Fangst av villrein på Hardangervidda. this case most likely Bergen. Punsvik, T. & Frøstrup, J. C. (red.). Fjellets nomade: Villreinen: Biologi – Historie – Forvaltning. Arendal. * Blehr, O., 2012. Medieval Reindeer Drives at Sumtan- gen, Hardangervidda: Two Interpretations. Forn- vännen 107:115–122. Otto Blehr and I have known each other for 50 — 2019. Sumtangen, a medieval communal reindeer years, from the early 1970’s when we both were (Rangifer tarandus) drive locality on the mountain researchers in the Hardangerviddaprosjektet for plateau Hardangervidda in Norway, once more. TverrvitenskapeligKulturforskning (HTK), he as an Fornvännen 114:242–246. ethnologist, I as an archaeologist. We were often Christie, W. F. K., Benhougene i Eyfjord, Ubb-ms-0222- of different opinions and disagreed on scientific XIII [Manuscript in The University of Bergen Library]. questions, the way it should be amongst dedicated — Fund af Been i Simedal, Eyfjord Præstegjeld i Hardan- researchers. However, in his last two debating ger, Ubb-ms-0222-XIV [Manuscript in The Uni- articles in Fornvännen 2012 and 2019, his scien- versity of Bergen Library]. tific arguments are at times overshadowed by an — Beenhoug i Eyfjord, Ubb-ms-0222-XV [Manuscript attitude that has little to do with serious research. in The University of Bergen Library]. As shown above, many of his verbal assaults Haukenæs, Th. S., 1884. Natur, Folkeliv og Folketro i Har- danger, belyst ved Natur og Folkelivsskoldringer, Even- against me are based on false claims, weak argu- tyr, Sagn, Fortællinger o.s.v. fra ældre og nyere Tid: ments, and he finds my articles “far too rich in Første del, Eidfjord. Hardanger. Nytt opptrykk 2013. unsubstantiated statements” (Blehr 2019, p. 245). Herteig, A., 1969. Kongers havn og handels sete: Fra de

Fornvännen 115 (2020) 50 Debatt

arkeologiske undersøkelsene på Bryggen i Bergen 1955– Nicolaysen, N., 1861. Reiseberetning, indsendt til det 68. Oslo. akademiske Kollegium. Foreningen til Norske For- Hufthammer, A. K., Bratbak, O. F. & Indrelid, S., tidsmindesmerkers Bevaring: Aarsberetning for 1860. 2011. A study of bone remains and butchering pat- Christiania. terns from mdieval mass-hunting of reindeer in Røed, K. H. & Hansen, G., 2015. DNA from ancient the South Norwegian mountain districts. Quater- reindeer antler as marker for transport routes and nary International, vol. 238:55–62. movement of craftspeople, raw material and prod- Indrelid, S., 2010. Om reinsdyrfangsten på Sumtangen ucts in medieval Scandinavia. Indrelid, S., Hjelle, i gamal tid. Villreinen 2010:28–34. K. & Stene, K. (eds.). Exploitation of outfied resources. Indrelid, S., 2013. «Industrial» reindeer hunting in Bergen. the south Norwegian mountains in the Viking Age Røed, K. H., Flagstad, Ø, Bjørnstad, G. & Huftham- and Early Middle Ages. Grimm, O. & Schmölcke, mer, A. K., 2011. Elucidating the ancestry of do- O. (eds.). Hunting in northern Europe until 1500 AD. mestic reindeer from the ancient DNA approaches. Schriften des archäologischen Landesmuseums, Quaternary International, vol. 238:83– 88. Ergänzungs-reihe 7. Neumünster. Skogland, T., 1993. Villreinens bruk av Hardangervidda. Indrelid, S., 2014. Oppdagelser på Hardangervidda. Kvinn- NINA Oppdragsmelding 245. Norsk Institutt for herad. Naturforskning. Trondheim. Indrelid, S., 2015. Medieval reindeer trapping at the Skogland, T. & Jordhøy, P., 1988. Konsekvensvurderingfor Hardangervidda mountain plateau. Indrelid, S., villrein ved omlegging av Bergensbanen på strekningen Hjelle, K. & Stene, K. (eds.). Exploitation of outfied Tunga-Låghellervatn,Ulvikkommune,Hord alandfylke . resources. Bergen. DN rapport 2/1988. Reguleringsundersøkelsene, Indrelid, S., Hufthammer, A. K., & Røed, K., 2007. Direktoratet for Naturforvaltning. Fangstanlegget på Sumtangen, Hardangervidda – Takken-Beijersbergen, L. M., 2017. Heaps of fascinat- utforskningen gjennom 165 år. Viking, vol. LXX: ing fragments: Reconstructing a medieval rein- 125–154. deer population based on bone middens from the Indrelid, S., & Hufthammer, A. K., 2011. Medieval Hardangervidda high mountain plateau, Norway. mass trapping of reindeer at the Hardangervidda Acta Zoologica, 98:340–361. mountain plateau, South Norway. Quaternary Inter- White, T. E., 1953. A method of calculating the dietary national, vol. 238:44–54. percentage of various food animals utilized by abo- Indrelid, S., Hufthammer, A. K., & Nesje, A. 2015. Om riginal peoples. American Antiquity, 38:266–278. reinsdyrfangst, rein og klima på Hardangervidda i eldre tid. Årbok for Universitetsmuseet i Bergen 2015. Bergen. Svein Indrelid Kjos-Hanssen, O., 1973. Reindeer Antlers and What Kalfarveien 74A They Can Tell Us about the Reindeer Population. NO-5022 Bergen, Norge Norwegian Archaeological Review, vol. 6, no. 2:74– 78. [email protected]

Fornvännen 115 (2020)